Formulating an Adaptation Strategy Documents/General/apf technical paper0… · Formulating an Adaptation Strategy 8 ... Moussa Cissé12, Mohamed El Raey13, Ulka Kelkar11,Jyoti K

Formulating an Adaptation Strategy

8

ISABELLE NIANG-DIOP1 AND HENK BOSCH2

Contributing AuthorsIan Burton3, Shaheen Rafi Khan 4, Bo Lim 5, Nicole North 6, Joel Smith 7,Erika Spanger-Siegfried 8

ReviewersMozaharul Alam 9, Anne Arquit Niederberger10, Suruchi Bhawal11, Moussa Cissé12,Mohamed El Raey13, Ulka Kelkar11, Jyoti K. Parikh14, Hubert E. Meena15, MohanMunasinghe16, Atiq Rahman 9, Roland P. Rodts17, Samir Safi18, Juan-Pedro SearleSolar19, Barry Smit20, and Thomas J. Wilbanks21

1 Department of Geology, Faculty of Science, University Cheikh Anta Diop, Dakar, Senegal2 Government Support Group for Energy and Environment, The Hague, The Netherlands 3 University of Toronto, Toronto, Canada4 Sustainable Development Policy Institute, Islamabad, Pakistan5 United Nations Development Programme – Global Environment Facility, New York, United States6 INFRAS, Zurich, Switzerland7 Stratus Consulting Inc., Boulder, United States8 Stockholm Environment Institute, Boston, United States9 Bangladesh Centre for Advanced Studies, Dhaka, Bangladesh10 Policy Solutions, Hoboken, United States11 The Energy and Resources Institute, New Delhi, India12 ENDA Tiers Monde, Dakar, Senegal13 University of Alexandria, Alexandria, Egypt14 Indira Gandhi Institute of Development Research, Mumbai, India15 The Centre for Energy, Environment, Science & Technology, Dar Es Salaam, Tanzania16 Munasinghe Institute for Development, Colombo, Sri Lanka17 Independent consultant, Ouderkerk a/d Ijssel, The Netherlands18 Lebanese University, Faculty of Sciences II, Beirut, Lebanon19 Comisión Nacional Del Medio Ambiente, Santiago, Chile20 University of Guelph, Guelph, Canada21 Oak Ridge National Laboratory, Oak Ridge, United States

8.1. Introduction 185

8.2. Relationship to the Adaptation Policy Framework as a whole 186

8.3. Key concepts 186

8.4. Guidance on formulation of an adaptation strategy 1878.4.1. Synthesise outputs of previous Adaptation

Policy Framework Components and other studies 187

8.4.2. Design the adaptation strategy: broad principles and considerations 189

8.4.3. Formulate options for adaptation policies and measures 190

8.4.4. Prioritise and select adaptation policies and measures 193

8.4.5. Formulate the adaptation strategy 196

8.5. Conclusions 196

References 197

Annex A.8.1. Methods for the prioritisation and selection of adaptation policies and measures 198

A8.1.1. Cost-benefit analysis 198A.8.1.2. Cost-effectiveness analysis 198A.8.1.3. Multi-criteria analysis 198A.8.1.4. Costing 199A.8.1.5. Using an expert panel 201A.8.1.6. Handling uncertainty and risk 201

Annex A.8.2. Linking climate change and sustainable development policies 201

Annex A.8.3. A hypothetical example of the use of multi-criteria analysis 203

Annex A.8.4. Useful internet addresses 204

CONTENTS

185Technical Paper 8: Formulating an Adaptation Strategy

8.1. Introduction

Adaptation to climate change is a process by which strategiesto moderate, cope with and take advantage of the conse-quences of climate events are developed and implemented(IPCC, 2001). Governments and/or communities can proac-tively adapt (Smit et al., 2001) but, at this time, the public pol-icy world is not yet motivated by climate change impacts.Policy making is dominated by competing priorities andinterest groups, and decisions with their own schedules –often unrelated to climate change. Among the many examplesare elections, natural disasters and fiscal crises. At the sametime, adaptation itself is a long-term process, requiring sus-tained attention. Developing an adaptation strategy is not asimple “one-shot” deal; instead it is an iterative, continuouslearning process. The objective of this Technical Paper (TP) isto help countries develop their own adaptation strategy byproviding guidance on setting priorities.

In designing an adaptation strategy, teams can acknowledge thereality of how a policy is made and yet, in the interests of cli-mate change adaptation, offer a clear vision of where it oughtto go. The strategy itself involves working within the contextand opportunities of the political structure, taking advantage ofopportunities as they arise, but having a sense of prioritiesbased on that vision.

This TP outlines the elements of adaptation strategies that arelikely to be consistent across the range of environmental or cli-mate contexts. The first activity is the synthesis of availableinformation. Once this data is compiled, the second activityinvolves designing an adaptation strategy with consideration toobjectives, indicators, and integration in national developmentplans and other synergies. The third activity involves the for-mulating the adaptation options for policies and measures.Selecting and prioritising the policies and measures, and thenextending the analysis beyond a simple list, comprise the fourthelement. Finally, the last activity is the formulation of an adap-tation strategy for implementation.

An adaptation strategy will, in many ways, be a “living” docu-ment; the process will not end with developing the strategy.Instead, it will mark the beginning of a new phase in whichlessons from implementing the approach are fed back into thestrategy to improve it over time (TP9). This strategy shouldinclude flexibility mechanisms to address the climate “surpris-es” that will almost certainly occur in the future, and accountfor new technologies and findings in the field of climate change(Klein et al., 1999).

The following pages focus on adaptation at a national level.However adaptation strategies, policies and measures must

Continuing the Adaptation Process

Use

rʼs

Gui

debo

ok

Assessing Vulnerability for Climate Adaptation

Formulating an Adaptation Strategy—Outlines how to formulate an adaptation strategy,policies, and measures, building on APF Components1-3and continuing into APF Component 5.

Assessing Current Climate Risks

Assessing Future Climate Risks

Assessing Current and Changing Socio-economic Conditions

Scoping and Designing an Adaptation ProjectScoping anddesigning an

adaptation project

Assessing currentvulnerability

Assessing futureclimate risks

Formulating anadaptation strategy

Continuing theadaptation process

TECHNICAL PAPERSAPF COMPONENTS

Ass

essi

ng a

nd E

nhan

cing

Ada

ptiv

e C

apac

ity

Eng

agin

g St

akeh

olde

rs

Figure 8-1: Technical Paper 8 supports Component 4 in the Adaptation Policy Framework

also be considered at the sub-regional and regional levels (e.g.,when they are related to shared resources, as is the case withinternational rivers). Furthermore, this TP focuses on publicand private planned – as opposed to autonomous – adaptation.1

The following sections describe the relationship of TP8 withthe Adaptation Policy Framework (APF) as a whole (Section8.2) and present key concepts (Section 8.3). Section 8.4offers guidance on formulating an adaptation strategy. Tasksdescribed here must be considered as indicative, since someof them may already have been completed. The annexesaddress methodological issues related to the process ofdeveloping an adaptation strategy.

8.2. Relationship to the Adaptation Policy Frameworkas a whole

This TP builds on the first three Components of the APFprocess (Figure 8-1) and provides direct input to Component 4.At this stage it is assumed that the team has designed andscoped the APF project, chosen an approach (Component 1),and analysed both the country’s adaptive capacity and the situ-ation that gives rise to a need for adaptation to climate change(Components 2-3). The analysis may have taken an approachthat emphasises vulnerabilities, climate risks, adaptive capaci-ty, and/or policies and measures.

The broad objectives of the strategy will be tailored to thecountry and its decision-making process, stakeholder consider-ations, the particular vulnerabilities and climate risks beingaddressed, and the resources available (adaptive capacity).These efforts provide the basis for developing and implement-ing an adaptation strategy for subsequent monitoring and eval-uation (Component 5).

8.3. Key concepts

A list of brief definitions of terms used throughout the APF canbe found in the Glossary. For important concepts specific tothis TP, more complete definitions are described in this section.These key concepts are the overall adaptation strategy, policies,measures, and the time horizons. The no-regret and low-regretoptions are also briefly defined, as well as the top-down andbottom-up approaches.

The adaptation strategy for a country refers to a general planof action for addressing the impacts of climate change, includ-ing climate variability and extremes. It will include a mix ofpolicies and measures with the overarching objective of reduc-ing the country’s vulnerability. Depending on the circum-stances, the strategy can be comprehensive at a national level,

addressing adaptation across sectors, regions and vulnerablepopulations, or it can be more limited, focusing on just one ortwo sectors or regions. In the Least Developed Countries, aNational Adaptation Programme of Action (NAPA)2 could wellbe developed into an adaptation strategy using the APF.

Generally speaking, policies refer to objectives, together withthe means of implementation. In an adaptation context, a poli-cy objective might be drawn from the overall policy goals ofthe country – for instance, the maintenance or strengthening offood security. Ways to achieve this objective may include:farmer advice and information services, agricultural researchand development, seasonal climate forecasting, and subsidiesor incentives for development of irrigation systems.

Measures are focused actions aimed at specific issues. Measurescan be individual interventions or they can consist of packagesof related measures. Specific measures might include actionswhich promote the chosen policy direction such as implement-ing an irrigation project, setting up a farmer information, adviceand early warning programme, developing a new scheme forcrop insurance, establishing a system of grain storage to protectagainst drought or crop failures, or providing financial incen-tive(s) to grow a specific crop, etc. Each of these measureswould contribute to the national goal of food security.

When defining these concepts, their distinctions are not alwaysclear. An example of the relationship among these terms follows.In Holland, the strategy to avoid rivers overflowing dikes isreferred to as “giving water more room”. A change in buildingcodes forbidding building or constructing any obstacle in theriverbed is one of the policies intended to carry out the strategy.Deepening and widening riverbeds and appointing overflow areasare specific measures. The following are examples of either poli-cies or measures: water conservation, investments in agriculturalinfrastructure, including roads to markets, drought and flood con-trol or alleviation measures, crop diversification, and alternativeoff-farm employment in rural areas. Many policies and measuresrelevant to adaptation may already exist in countries.

Setting of time horizons is needed when defining a strategy, pol-icy, or measure, and also for monitoring the implementation ofan adaptation strategy (TP9). Generally, strategies would belong-term in nature, and policies targeted at the medium- tolong-term. Measures may have an implementation time of anylength, but are expected to have sustained effects. Prioritisation,mostly of measures, but in some cases also of (alternative) poli-cies, will take the whole period into account. A NAPA, which ismeant to communicate the most urgent needs of LeastDeveloped Countries, is likely to contain measures with a shortimplementation time, but with immediate and, preferably, long-term effects. A further discussion on planning and policy hori-zons can be found in TP5.

Technical Paper 8: Formulating an Adaptation Strategy186

1 Autonomous adaptation refers to adaptation action taken by individuals or systems without involvement by government. While not considered in this TP,autonomous adaptation may lead to ideas for government-assisted adaptation and is an important determinant of the “adaptation environment”.

2 NAPAs are a quick participatory vulnerability and adaptation process through which Least Developed Countries will be able to develop project proposals relative to their immediate and urgent adaptation needs addressing the most vulnerable areas, systems (UNFCCC Decision 28/CP7).


No-regret options are measures or activities that will proveworthwhile even if no (further) climate change would occur.Low-regret options are no-regret options that require smalladditional outlays to cater to the negative effects of climatechange. The notion of “no regret” or “low regret” may be use-ful when trying to obtain (outside) finance (Box 8-1).

The notions of top-down and bottom-up generally apply to thecontext of planning. Top-down planning would typicallyemanate from higher levels in government such as planning min-istries. A bottom-up planning process would start at the locallevel and, from there, progress “up” to the decision-makers. TheAPF places equal importance on the bottom-up approach inorder to take account of existing adaptation and coping mecha-nisms at the local level. Effective planning processes may bethose that successfully integrate the two approaches. For bothapproaches, stakeholder involvement is equally important.

8.4. Guidance on formulation of an adaptation strategy

Adaptation teams can choose to use a top-down or bottom-upapproach for formulating an adaptation strategy, policies andmeasures. While the bottom-up approach plays an importantrole throughout the APF, this TP emphasises the top-downapproach and assumes the highest level of government buy-in.In practice, the selection of the approach means choosing thepath of least resistance and taking into account the cen-tralised/decentralised nature of decision-making in that country.

With the top-down approach, the overall policy direction of thecountry will guide the design of the adaptation strategy. In turn,the objectives of the strategy will guide the selection, design,and implementation of new policies and measures – a processthat requires a high level of political will. At the same time, anadaptation strategy can assist a country in meeting existingobligations to international agreements – e.g., in biodiversityand desertification. Development or revision of national plans,such as those for drought management, coastal management,biodiversity conservation, and forest management, can presentadded opportunities for integrating climate change concernsinto related planning processes. Given the opportunities forsynergies and policy coherence, all of these imply a top-downapproach. But a top-down approach requires significant capac-ity for policy design (TP1) and can be more ambitious than itsalternative. In circumstances where this capacity is limited, abottom-up approach may be more pragmatic.

A project team may prefer to begin the APF process with a setof articulated measures and/or policies that have already beenanalysed in terms of vulnerability, climate risks, future socio-economic conditions, and adaptive capacity. In this case, a bot-tom-up approach to developing the adaptation strategy – i.e.,formulating the strategy that coheres with the measures – isobviously more appropriate. An example of a bottom-upapproach can also be found in Klein et al. (1999), where poli-cies and measures are first designed, then embedded in anadaptation strategy. Generally, these policies and measures

have been developed for reasons other than climate change,and climate change is included as an additional consideration.

This TP presents the five different activities involved in formu-lating an adaptation strategy (Figure 8-2). Depending onnational circumstances, some of these tasks may already havebeen performed, mainly through Initial National Communica-tions to the UNFCCC and NAPAs:

1. Synthesise outputs of previous APF Components andother studies

2. Design the adaptation strategy3. Formulate adaptation options for policies and measures4. Prioritise and select adaptation policies and measures5. Formulate an adaptation strategy

While this paper emphasises the use of prioritisation method-ologies, the APF recognises the value of expert judgement in policy-making processes when applied rigorously. Suchalternative approaches will be presented in related APF publications.

8.4.1. Synthesise outputs of previous Adaptation PolicyFramework Components and other studies

The first activity is to synthesise what is known about the coun-try’s vulnerability and adaptive capacity with regard to thepotential impacts of climate change. Within the APF framework,this will involve assessing the adaptation options identified byprevious Components – namely, the outputs of the project scop-ing process, as well as the current and future vulnerabilityassessments. If sufficient information exists, some users maychoose to begin with Component 4, by applying this paper.

Most likely, this information will include responses around thefollowing questions:

• What ecosystems, sectors, regions, and populationsare particularly vulnerable to climate change? (APFComponents 2 and 3; TPs 3- 6).

• What is the current level of adaptation and adaptivecapacity that constitutes the adaptation baseline? (APFComponent 2; TPs 3 and 7). The determinants of adap-tive capacity may include technological advances, insti-tutional arrangements, new and existing policies, avail-ability of financing, level of information exchange, etc.

• What sets of indicators were chosen for vulnerability,socio-economic and adaptive capacity analyses? (TPs3, 6, and 7). These should lay the basis for evaluatingand prioritising alternative adaptations.

• Which set of stakeholders represents the various sec-tors, regions, and populations being considered? (TP2).

Teams using the APF will want to revisit their initial analysis of thepolicy process within their country (Component 1; TP1), particu-larly in light of the current socio-economic context and the scenar-ios of potential future conditions they will have developed (TP6).


APF

Component 1: Scope project

Com

pone

nt 4

: F

orm

ulat

e an

ada

ptat

ion

stra

tegy

Component 5: Continue adaptation process

Component 2: Assess currentvulnerability

Component 3: Assess futureclimate risks

Task 1: Synthesise outputs ofprevious APF Components andother studies and collect potential adaptation options

Task 3: Formulate adaptation options forpolicies and measures • Description • Costs • Impacts • Barriers

Task 4: Select and prioritiseadaptation policies and measures • Country context/criteria • Determine and use tools for prioritisation/selection • Cross check results with stakeholders

Task 5: Formulate an adaptation strategy • Organise the adaptation options in a strategy • Check coherence and synergies with other MLA, and sustainable development • Consider institutional arrangements

Implementation

Existing lists of adaptation optionsin National Communications

Task 2: Design the adaptation strategy

Monitoring and evaluation of the strategy

NAPAs

Pilot projects

Indicators

Inclusion in national development plans

Projects for funding

Figure 8-2: Component four in the Adaptation Policy Framework context


As stressed throughout the TPs, the APF process is expected to beembedded within the context of current policies and to build uponearlier national assessments. These assessments include, for exam-ple, the National Communications to the UNFCCC, NAPAs, pre-vious vulnerability and adaptation studies, and pilot projects or pro-grammes. NAPAs will likely only address a few priority sectors andsystems, and the analysis may need to be extended to other systems.Guidance from the Intergovernmental Panel on Climate Change(IPCC) 3 may provide additional ideas.

Upon completing this task, the team should have a clear idea ofthe vulnerability of the country with regard to climate change,the capacity of the country to adapt, and the overall policy con-text to be considered while developing the adaptation strategy.A detailed plan to engage stakeholders in strategy developmentshould be also prepared, as well as indicators that could be usedto assess alternative adaptation strategies.

8.4.2. Design the adaptation strategy: broad principlesand considerations

The adaptation strategy consists of a broad plan of action to beimplemented through policies and measures over the short-,medium- and long-term. The objectives of the adaptation strategycan be very specific (e.g., reduce the vulnerability of a sector), orquite broad (e.g., reduce poverty, achieve the MillenniumDevelopment Goals, etc.). From overall objectives, more specificgoals can be derived. In any case, stakeholders, including govern-ment, will define these objectives (TP2) under Component 1 ofthe APF process.

An adaptation strategy is best supported by a set of instruments

designed collectively. Without regulatory and economic instru-ments, adaptation to climate change will remain at the level ofeducation and awareness raising. Many economic and espe-cially regulatory instruments, however, may not work effec-tively without enforcement and compliance. A package of poli-cies and measures should be designed to complement and rein-force each other. Policy instruments can be selected using bothformal and informal methodologies in decision making, includ-ing decision-support tools (described later in the text).

A core set of policy instruments for implementing strategicdecisions can be incentive-based or “control and command”interventions. The selection of instruments is closely linked tothe socio-economic analysis, in which the barriers that impedeadaptation have been identified. TP9 discusses common barri-ers to implementation in the context of environmental gover-nance. Common types of policy instruments are outlined inBox 8-2.

As mentioned above, an adaptation strategy should also achievesynergy with other environmental strategies. Climate changeissues are closely linked to the Convention on Biological Diver-sity as well as to the Convention to Combat Desertification. Forexample, drought early warning systems and contingency plans,food security systems, the development of alternative livelihoodprojects or sustainable irrigation programmes for both crops andlivestock each could be considered an adaptation option in aridand semi-arid areas. At the same time, each of these could serveas a Component of a National Action Plan to combat desertifi-cation and drought in these same areas. To ensure greater effi-ciency and enhance the impact of all strategies, the APF teamshould ensure synergy among the responses to these differentRio Conventions.4

Box 8-2: Common types of policy instruments

• Legislative, regulatory, and juridical instruments. Legal instruments set limits and provide sanctions, but can bedifficult to enforce. Examples are: laws, by-laws, regulations, standards, constitutional guarantees, and nationalagreements based on international conventions.

• Financial and market instruments. Fiscal instruments can influence behaviour by sending price signals. They area powerful set of instruments for raising revenue for environmental management, but tend to be difficult to imple-ment politically. Examples of market-based approaches are: property-rights based approaches (concessions,licences, permits), price-based approaches (taxes, payments for amenities, user fees, tax credits for investmentfunds, performance bonds), perverse subsidy removal, and market-based measures (labelling, procurement poli-cies, product certification, information disclosure requirements).

• Education and informational instruments. Education instruments raise awareness, and over time, they change soci-etal values. Examples are: consumer information, public awareness campaigns, and professional development.

• Institutional instruments. Private companies, corporations, and communities often adopt such policy instruments.Examples are: environmental management systems, management policies and procedures for service contracts.

3 Especially the 1995 and 2001 Working Group II Reports (Watson et al., 1996; McCarthy et al., 2001).4 As it is expressed under point 39c of the Plan of Implementation adopted by the World Summit on Sustainable Development.


The strategic planning process will require “cross-sectoral co-operation, an interdisciplinary approach and considerable politi-cal will” (Least Developed Countries Expert Group, 2001;Annex A, OECD, 2002). It will need, among other things,engagement of each of the ministries responsible for develop-ment planning in the country. An adaptation strategy may con-tain several objectives (Box 8-3).

Based on the key priority systems identified earlier in the APFprocess, indicators will have been defined that can be used toassess the success, difficulties, failure of the adaptation strate-gy once implemented. Some of these indicators (vulnerabilityand socio-economic) are discussed respectively in TPs 3 and 6.

8.4.3. Formulate options for adaptation policies and measures

Once the broad objectives of the adaptation strategy have beendetermined, it is possible to formulate policies and measures toachieve these objectives. If included at this stage of the process,several factors will facilitate integration of adaptation policiesand measures later on.

• An important step in the process of formulating options isthe integration of adaptation policies and measures betweendifferent sectors – and with existing policies and measures.This step builds on the synergies identified early on in strat-egy design by ensuring that the overlaps and intersectionsbetween adaptations and existing policies and measures areco-ordinated to the benefit of both. With integration, poten-tial conflicts between adaptations in different sectors, andbetween proposed adaptations and existing policies andmeasures, can be avoided or limited. It is well known, forexample, that developing an adaptation strategy in the agri-culture sector without considering the water sector is notreally feasible due to the relationship between the two. Thesame can be said for human health and water. Even purelystructural adaptations (e.g., seawall construction, changes in

agricultural practices, establishment of early warning sys-tems), therefore, will require integration. Of course, inte-grating adaptations leads to the issue of common (shared)costs and benefits, and the problem of how to attribute thosecosts and benefits to the different sectors/projects. An exam-ple of integration is given in Box 8-4.

• An evaluation of the relevant sectoral policies is therefore rec-ommended. For example, a country’s agricultural policy mayhave development objectives that are threatened by climatechange. (TPs 1 and 6 discuss the adaptation baseline.) Suchdevelopment objectives may include maintenance or strength-ening of food security, the promotion of commercial crops forexport or the production of crops that serve as industrial rawmaterials or substitute for imports. To achieve these objec-tives, adaptation policies may require additional effort for cre-ation and/or improvement of all or some of the following:farmer advice and information services, agricultural researchand development, seasonal climate forecasting, taxes and/orsubsidies or incentives, irrigation, water conservation, invest-ments in agricultural infrastructure including roads to mar-kets, drought and flood control or alleviation measures, cropdiversification, alternative rural off-farm employment and soon. Many different policy mixes are possible, and an actualpolicy usually tries to satisfy as many objectives as possible.Where the objectives are found to be incompatible or in con-flict, an assessment of the trade-offs is necessary.

• Specific measures can be developed to support the chosenpolicy direction, e.g., an irrigation project, a farmer infor-mation, advice and warning programme, new scheme forcrop insurance, a system of grain storage to be held as pro-tection against drought or crop failures, financial incentivesto grow a specific crop.*

• Adaptation options can be considered at different timescales (Parry and Carter, 1998), as some will have longer-term policy impacts than others. This factor may influencehow urgently these policies and measures need to be

Box 8-3: Five generic objectives of adaptation to climate variability and change

1. Increasing robustness of infrastructure designs and long-term investments – e.g., by extending the range of temper-ature or precipitation a system can withstand without failure and changing the tolerance of loss or failure;

2. Increasing the flexibility of vulnerable managed systems – e.g., by allowing mid-term adjustments (including changesof activities or location) and/or reducing economic lifetimes (including increasing depreciation);

3. Enhancing the adaptability of vulnerable natural systems – e.g., by reducing other (non-climatic) stresses andremoving barriers to migration (including establishing eco-corridors);

4. Reversing trends that increase vulnerability (also termed “maladaptation”) – e.g., by introducing setback lines fordevelopment in vulnerable areas, such as floodplains and coastal zones;

5. Improving societal awareness and preparedness – e.g., by informing the public of the risks and possible conse-quences of climate change and setting up early-warning systems.

Source: Klein and Tol (1997)

* Many developing countries may need to consider low-technology options to overcome difficulties arising with the maintenance of “hard” adaptation mea-sures. Experience with imported technologies indicates that it is not always wise to simply “paste” new technologies in the context of developing countries.


implemented over the whole planning horizon of the adap-tation strategy. Examples are:

• long-term adaptations that are responding to meanchanges in climate (river basin planning, institutionalchanges for water allocation, education and research);

• tactical adaptations concerned with mid-term consid-erations of climate variability (flood-proofing, waterconservation measures);

• contingency adaptation related to short-term extremesassociated with climate variability (emergency droughtmanagement, flood forecasting);

• analytical adaptations considering climate effects at allscales (data acquisition, water management modelling).

• To ensure that the adaptations identified are suitable to thechallenge, it is important to engage stakeholders that canprovide perspective on the feasibility of proposed options.

Box 8-5 gives examples of adaptations. Other categories ofadaptation are organised in Smit et al. (2001) by functional out-come, type of policy instrument, and level of application.Sectoral measures are also available in the programmes ofinternational organisations, such as the Food and AgriculturalOrganisation, government ministries and technical depart-ments, research centres and non-governmental organisations,etc. Different adaptation measures have also been described inexisting adaptation planning guidebooks and reports.5

Many adaptations will have been identified in the previousAPF Components (especially Component 2; TPs 3, 4 and 6).Such adaptation may be currently in place to address climatevariability (e.g., interannual variability of precipitation,ENSO) and extremes, such as droughts, floods and cyclones.Many of these practices are developed by local communitiesand especially by highly vulnerable people, existing on themargins of society. Such adaptations need to be consideredwhen the strategy is developed, not only because these mea-sures have been tested in the field, but also because they aremore likely to be accepted by the communities. The teamshould develop not only lists of adaptations but also include

assessments of their experiences (what has worked and whathas not?) in order to develop new and revised adaptation poli-cies and measures (Figure 8-3). Other adaptations may bededuced from the analysis of future climate risks (APFComponent 3 and TP5), and obtained from the literature,research centres, and clearinghouses (e.g., transferring avail-able technologies at an international level). Experiences fromadaptation policies and measures implemented in other coun-tries could also serve to illuminate.

Figure 8-3 outlines the generic process involved in identifyingand assessing adaptations. The first step involves the identifi-cation of existing and potential adaptations. The second stepreviews these options in light of their actual or potential effec-tiveness in addressing current climate vulnerability and risk.The next step involves an assessment of the effectiveness ofthese options in light of potential climate futures. The fourthand final step involves prioritising certain adaptations over oth-ers, based on agreed criteria.

Once identified, adaptations have to be formulated in such amanner that their selection and prioritisation is possible usingvarious methods. Since options will vary widely, it is only pos-sible to outline typical information requirements, rather thangive a prescribed format. Typical requirements are:

• Description of the measure, indicating objective(s),location (e.g., international, regional, national, sub-national, or local), timing of and responsibilities forimplementation, and financing. This description wouldaddress the technical feasibility of measures, barriers totheir implementation (e.g., cultural, social), the capaci-ty to implement and sustain the measure, the culturalacceptability of the technology involved, etc.

• Estimated costs of the measure. The cost is a prerequi-site for ranking a measure and including it in the(national, provincial, etc.) budget, or in a wider adap-tation programme. Costs could be a one-time expen-diture for capital investments, and recurrent costs (e.g.,in the case of certain public health campaigns), includ-

Box 8-4: Considering sea-level rise in reconstruction of seawalls in Belize

In 1998 Hurricane Mitch stalled offshore of Belize for several days. This powerful hurricane generated huge waves thatslammed onto Belize’s coast and generated unusually high tides. Over 90% of the country’s piers were destroyed. Therewas tremendous erosion and mechanical damage to the reef and coastal infrastructure, including the seawall in Belize City.The following year the government embarked on a project to rebuild parts of the seawall. Belize was in the process ofpreparing its Initial National Communication, and the project co-ordinator wrote to the Ministry of Works, which had com-missioned the construction of the seawall, explaining the projected increase in sea level expected in the coming decades.The project co-ordinator advised the Ministry to either consider making the seawall high enough to retain a higher sea level,or to build it in such a way that it could be raised in the future. The chief engineer agreed to factor in climate change andthe seawall is higher than originally planned.

5 Such as the UNEP handbook for V&A studies (Feenstra et al., 1998), the US Country Studies Programme guidebook (Benioff et al., 1996), and the IPCCreports on impacts and adaptation (Watson et al., 1996; McCarthy et al., 2001).


Box 8-5: Types of adaptation measures

Adaptation measures may be grouped according to whether they are sectoral (e.g., introduction of improved agriculturalvarieties), multi-sectoral (e.g., use of improved watershed and coastal zone management methods), or cross-sectoral (e.g.,promotion of public awareness, climate research, and data collection).

• Sectoral measures relate to specific adaptations for sectors that could be affected by climate change. In agricul-ture, for example, reduced rainfall and higher evaporation may call for the extension of irrigation. For infrastruc-ture, sea level rise may necessitate improved coastal protection or relocation of population and economic activi-ties. In most cases, measures will mean a strengthening of existing policies, emphasising the importance of basingclimate change policies on existing coping mechanisms and the necessity of integrating them into national devel-opment plans.

• Multi-sectoral measures relate to the management of natural resources that span sectors – e.g., water managementor river basin management. Integrated coastal zone management is also considered an appropriate framework toconsider technical adaptation measures such as dike building, beach nourishment, etc. (Bernthal et al., 1990). Theecosystem approach to climate change adaptation involves the integrated management of land, water and otherresources that promotes their conservation and sustainable use in an equitable way (Orlando and Klein, 2000).

• Cross-sectoral measures can span several sectors and include the following:

Education and training: Introduction of climate change issues at different levels of the educational system is anongoing process that can help to build capacity among stakeholders to support adaptation in the future, and canhelp to develop appropriate research activities and a greater awareness among citizens.

Public awareness campaigns: Such campaigns can raise awareness and disseminate information in order toincrease the concern and involvement of the broad array of stakeholders. These campaigns can also be an oppor-tunity for adaptation decision makers to better understand the perception and views of the public on climate changeand adaptation.

Strengthening/changes in the fiscal sector: Public policies may encourage and support adaptation of individualsand the private sector, particularly through the establishment of fiscal incentives or subsidies.

Risk/disaster management measures: These measures include the development of early warning systems, in par-ticular for extreme events like cyclones (that can be predicted only a few hours before), and for droughts, floods,El Niño-Southern Oscillation (ENSO) (that can be predicted several months before). Emergency plans, extremeevents relief and recovery measures also belong to this type of measure. Generally, the success of these measuresdepends upon good communication systems and a certain level of trust among users.

Science, research and development (R&D) and technological innovations: R&D and innovation are needed to enableresponses to climate change in general, and to enable specific responses to climate change vulnerability, includingeconomic valuation of adaptations, technological adaptations (development of drought- or salt-resistant crop vari-eties), and investigations of new sources of groundwater and better resource management. It may also be necessaryto adapt existing technologies to fit with the adaptation demands – e.g., the development of more energy-efficient airconditioning systems, low-cost desalination plants, and new technologies to combat saltwater intrusion.

Monitoring, observation and communication systems: These systems may have to be created or strengthened, par-ticularly for climate-related parameters, but also for other indicators of climate change and impacts (e.g., sea-levelrise, changes in species composition of ecosystems, modification of piezometric levels, etc.). This monitoring willallow policy-makers to adjust the adaptation strategy based on confirmed changes in the climate (TP9).


ing operational costs for project-type measures. Apartfrom direct costs, there are often indirect costs (e.g., inthe form of an additional burden to the administrativesystem of the country) and external costs (linked e.g.,to negative impacts in another sector). Costs should – tothe extent possible – be expressed in monetary form.When this is not possible – as may be the case, forinstance, in relation to changes in ecosystems – thesefactors have to be incorporated qualitatively. Methodshave been developed to successfully quantify and valuethe use of resources for which there is no market price(Annex A.8.1); such methods can be used in the formu-lation process.

• Estimated benefits of the measure. The impacts ofthe measures on the environment and on society canbe determined by comparing the “with” and “without”case.6 These impacts need to be described in terms oftheir contribution to the objectives or criteria, againpreferably in monetary terms. As is the case withcosts, impacts may be system specific (e.g., humanhealth, agriculture, environment, biodiversity, infra-structure, etc.), and be multi- or cross-sectoral. Costsand benefits are mirror images and often benefitsresult in reduction of the (social) costs. Examples arethe reduction of typhoon damages resulting frominstalling an early warning system and a reduction offlood damages with increased heights of dikes. Theevaluation of options should include equity consider-

ations, and an assessment of to whom benefits accrueis therefore needed.

The main output of this task is a portfolio of adaptation mea-sures and policies. The next task will be to select and prioritisethese options.7

8.4.4. Prioritise and select adaptation policies and measures

After adaptation policies and measures have been formulated,they can be prioritised with various methods and, subsequently,rejected, postponed, or selected for implementation. Given therange of climate change impacts and the measures to avoid or mit-igate these impacts, it is unlikely that one single method can han-dle all possible cases. From a methodological point of view, thethreats caused by climate change are not essentially different fromwhat people have been experiencing in the past. Therefore, eval-uation methods used in the selection and prioritisation exerciseneed not differ either. However, the increase in frequency andintensity of extreme events puts more emphasis on the treatmentof uncertainty and risk. Sensitivity and risk analysis are thereforevaluable elements in the decision-making process (Annex A.8.1).

Formal methods for prioritisation can most easily be applied toproject-type (sectoral, and multi-sectoral) adaptation measures. Inthe case of cross-sectoral measures, such as institutional reformand legislation, it is often difficult or impossible to quantify thebenefits or impacts of a measure. For these measures, it may then

Figure 8-3: Identification, analysis and prioritisation of adaptation options

6 The method of comparing “with” and “without” is not confined to cases where costs and/or benefits can be quantified and/or expressed in monetary values.In qualitative reasoning the concept helps to avoid including impacts and costs resulting from autonomous development. An often observed fallacy is that ofcomparing the “before” and “after” circumstances.

7 Many Initial National Communications to the UNFCCC included lists of adaptation options; however, these were not necessarily described and analysed(task 3) and prioritised (task 4) in a manner that could facilitate informed adaptation planning.

(Step 1) Identifying current adaptationoptions/policies for floodplain

Step 2: How do theseoptions and policiesneed to be improvedto deal with today’s

climate? Reduce vulnerability today?Additional strategies

identified?

Step 3: How do theseoptions and policiesneed to be improvedto deal with futureclimate change andvariability? Reducevulnerability in thefuture? Additional

strategies identified?

Step 4: How to priori-tise these adaptationoptions and policies?

How to integratethem into existing

policies? How to fillin the gaps?

Costs? Feasibility?Effectiveness?

i.e., Structural: levees, dikes

i.e., Near real time early warning system

i.e., Disaster managementpolicy/strategies

i.e., Poverty reduction strategies

(Step 1) Identifying potential adaptationoptions/strategies, policies

i.e., Relocation

New options? Populate list from Steps 2 and 3


be necessary to employ informal, qualitative and subjective waysto determine their attractiveness.

Four main methods are likely to be particularly useful to theprioritisation process. These are:

• Cost Benefit Analysis (CBA)• Cost Effectiveness Analysis (CEA) • Multi-Criteria Analysis (MCA)• Expert judgement

Box 8-6 presents the pros and cons of each method. TheCompendium of Decision Tools lists a number of additionalmethods, including sector-specific tools (UNFCCC, 1999). TheHandbook on Methods for Climate Change Impact Assessmentand Adaptation Strategies (Feenstra et al., 1998) discusses theselection issue in great detail.

The selection of a method to evaluate policies and measuresshould be based on the real-life situation of the country, includ-ing available data and resources, and on the requirements of theprospective financier of the measures (government, outsidefinancing). Formal methods such as CBA (discussed further inAnnex A.8.1) can best be applied if outside financing is requiredor if planning authorities in the country so demand. It is impor-tant at this stage that planners are involved, either directly or inwriting terms of reference for the studies. In many cases the pol-icy process of a country will involve expert and political analy-sis and judgement. If a plan is to be entered into the budget of thegovernment, an estimate of the cost will normally be required,with non-monetary elements such as institutional/organisationalcosts and cultural realities, and types of adaptive capacity need-ed for implementation quantified to the extent possible.

The flowchart in Figure 8-4 explains the reasoning presentedin Box 8-6 and – because it attaches great importance to theaccuracy of results – this chart applies especially in the laterstages of the assessment of adaptation policies and measures,i.e., just before they are ready to enter into the adaptationstrategy, the national development plan, or the national or sec-toral budgets.

For evaluation and ranking of measures, it will be necessaryto choose criteria to weigh the different concerns. These cri-teria can also act as indicators of the success or failure torealise the objectives, and can be used by a monitoring-eval-uation programme for the adaptation strategies, policies andmeasures (TP9). The adaptation measures and policies shouldbe evaluated within the same policy context as measures andpolicies introduced to alleviate poverty, or to foster economicdevelopment. The country’s policy context should be takeninto account when choosing criteria for the evaluation ofadaptation measures. A sample set of criteria is offered in theNAPA guidelines (GEF, 2002), as outlined in Box 8-7. TheNAPA Guidelines stress that the selection of criteria shouldbe a country-driven process and that the list of criteria is notmeant to be prescriptive. Other criteria that may be applicableare gender, sustainable development, equity, etc. Annex A.8.2suggests a way to explore the impact of various measures onsustainable development; the team may find it helpful toorganise data in that respect. In general, a useful way toorganise data on adaptation policies and measures is toexpress the effects of measures in an “impact matrix” inwhich the measures to be compared are explored against therelevant criteria (see Table A-8-2-1 in Annex A.8.2 for anexample).

Box 8-6: Four main methods for prioritising and selecting adaptation options

The four major methods used for prioritising and selecting adaptation options – cost benefit analysis, multi-criteria analysis,cost effectiveness analysis, and expert judgement – vary in a number of ways. Some of these are outlined here:

• CBA can handle optimisation and prioritisation; it also provides an absolute measure of desirability, albeit judgedby only one criterion, i.e., economic efficiency. CBA has comparatively heavy data requirements.

• MCA is suitable when more criteria are thought to be relevant, and when quantification and valuation in monetaryterms is not possible. MCA is normally used for the ranking of options. But if the “do-nothing” case is includedas an alternative, it can also help to clarify whether the measure is better than simply “bearing with the situation”.Subjective judgement plays an important role in this method, making outcomes more arbitrary than that of CBA.8

• CEA is a method that falls somewhere between CBA and MCA. As is the case with MCA, CEA only produces a ranking.• Expert judgement is a discipline in its own right and has its own place in the domain of policy making (Section 8.4).

Given that CBA is the more objective method and can handle optimisation, it may be the most desirable option. However,this depends on the purpose and stage of the analysis. In cases where important criteria cannot be accommodated in CBA(such as sociological or cultural barriers), or when benefits cannot be quantified and valued (such as the benefits of pre-serving biodiversity), MCA is preferred. If desired, the outcomes of CBA can be incorporated into MCA, making the over-all analysis a hybrid one.

8 For more detail on CBA and MCA (or decision analysis in general), and risk analysis, refer to Annex A.8.1 and to relevant textbooks (see the references section).


The process of prioritisation and selection of adaptations mustinvolve a wide array of stakeholders. Multi-criteria analysis maybe the best-known selection method involving stakeholders at thegrass-root level. It is important to stress that during all steps ofthe process (choice of the method, choice of criteria, use of themethod), stakeholders should be involved so that the process ofimplementation can be facilitated.

Given the uncertainties and the long time frame of climatechange impacts (TP5; Willows and Connell, 2003), the so-calledno-regret and low-regret adaptation options may be among themost attractive. The merits of an adaptation can also be comparedfor different climate scenarios, including the assumption that no(further) climate change will occur (TP6).

Box 8-7: Sample criteria for selecting adaptation options

As a NAPA may well precede a full-blown adaptation project, these criteria may be of use in both, and are therefore brieflydiscussed below:

• The expected level of damage is an indication of the benefits to be gained by preventing or mitigating this dam-age. Damage levels may need to be disaggregated into sector- or system-specific impacts, depending on the mea-sure and the situation in the country with regard to adaptive capacity, the health situation, food security, and so on.If this implies that this criterion has to be broken up into sector- or system-specific criteria, the appropriate evalu-ation method will often become multi-criteria analysis (Annexes A.8.1 and A.8.3).

• Poverty reduction will enhance adaptive capacity and could be a goal in itself (and then a criterion) or a “by-prod-uct” of a measure (e.g., adaptation in agriculture). The two criteria can easily overlap and lead to double counting.PRSPs provide information on (“autonomous”9) development. Important synergies may be found between effortsunder the PRSP and adaptation plans.

• Synergies with multilateral environmental agreements can have the form of cost savings or of additional ben-efit (e.g., the introduction of drought resistant crops which reduces desertification).

• Cost effectiveness (or just “costs”) is a main criterion at the same level as “impacts/benefits”.

Source: NAPA Guidelines (GEF, 2002)

9 Autonomous in the vulnerability and adaptation analysis.

One objective/criterion?Quantification and valuation possible?

Yes Yes

CBA

One objective, but benefits not in money terms

CEA

Two objectives, weightingof benefits possible?

Yes

MCANo data?Expert Panel

More objectives, some orall costs and benefits notin monetary units

No

No

No

Figure 8-4: Choosing a tool for the prioritisation and selection of adaptation options


At the end of this task, the team would have designed a set ofreasonable policy alternatives and have evaluated them interms of criteria and objectives. Usually no one option willbe superior in all respects. Option A may be more efficient,but less acceptable to some vested interests with politicalpower; Option B may be less efficient, but less vulnerable toclimate change in the medium- and longer-term. It is the taskof the prioritisation and selection analysis to give a rankingbased on explicit criteria and their weights. It is the up to thepolicy process to make the final choice. In other words, thepolicy analysis is subsidiary to the policy choice, which inthe final analysis is made at the political level. A good poli-cy process constrains the political choice to a limited set ofviable alternatives.

8.4.5. Formulate the adaptation strategy

Once the prioritisation process is completed, an adaptationstrategy can be prepared with a combination of different mea-sures and policies. The adaptation strategy consists of a plancontaining the collection of measures selected for implemen-tation, a time frame and other operational modalities forimplementation 10. This document should describe the scop-ing of the issues, identification of options, approaches takento examine and evaluate the options, and transparency of theassessment process.

An implementation plan can be developed in which policiesand measures are categorised according to:

1) How they are to be incorporated into existing sectoralstrategies, national development plans, poverty reduc-tion strategies, etc. (e.g., management plans, educa-tion and research programmes, laws to be developedor enforced)

2) Additional plans, policies, measures and/or projects thatspecifically address climate change that may be neededif gaps have been identified in the current policy frame-work. Some measures will likely require financing,either government or external, while others could betaken aboard within the regular national budget.

3) A further distinction could be made between urgentpolicies, measures and projects, and those that aresomewhat less urgent. Some of the measures may beimplemented right away, while others may requiredetailed feasibility studies.

During this formulation and adoption process, it is important toinclude stakeholders at all levels (national to local) to gain pub-lic acceptance of the strategy (TP2). The resulting strategy doc-ument must be formally recommended for adoption, whetherthrough a government decision, or through stakeholder consul-tation. TP9 suggests how adaptation strategies can be imple-mented, monitored and evaluated.

8.5. Conclusions

A key objective of the APF is to facilitate the development andimplementation of an adaptation strategy. However, formulat-ing a plan that is only motivated by climate change may beunrealistic, not only because adaptation involves different sec-tors, regions and populations that are vulnerable to climatechange, but because climate change is often far from being thefirst concern of most of decision-makers. Instead, decision-makers tend naturally to be more concerned with urgent goalssuch as poverty reduction and national development. It is thusof uppermost importance that, over the course of an adaptationproject, efforts are made to build an understanding among keystakeholders that adaptation to climate change may become anecessary undertaking to achieve these same objectives.

Box 8-8: No-regrets and low-regrets options

Given uncertainties and the long time frame of climate change impacts (TP5; Willows and Connell, 2003), two generaltypes of adaptation options discussed here may often be most appropriate and most readily funded:

• No-regrets: These are options that are justified by current climate conditions, and are further justified when climatechange is considered. For example, reducing water pollution could improve potable water supplies. The pollutionreductions may be more valuable should climate change reduce water supplies or degrade water quality. The samecan be said for introducing market reforms. However, an irrigation scheme for a drought-prone area may becomemore attractive when periods of drought, as a result of climate change, occur more often or become more severe.

• Low-regrets: Low regrets changes are those made because of climate change, but at a minimal cost. Thus, there is “lowregret” if the investment proves not to be needed under future climate conditions. For example, incorporating risks ofclimate change in design of infrastructure may offer improved protection against current extreme climate events, aswell as potential future events under climate change, while increasing costs only marginally (hence the “low” regret).

10 Sources of adaptation funding are expected to be provided under the UNFCCC through the GEF, such as the GEF Trust, Special Climate Change, and LeastDeveloped Country Funds, and bilateral funds and national budgets.


Some adaptation measures may already be in place in somecountries. This paper provides guidelines for developing anadaptation strategy by building on existing mechanisms, andshould not be seen as prescriptive. The main output of theComponent 4 is an adaptation strategy with an implementationplan for formal adoption. TP9 deals with both implementationand evaluation of adaptation. Since adaptation is a continuousprocess that needs to be informed regularly by evaluation of theadaptation strategy, implementation and monitoring are treatedas integral parts of the strategy development process.

References

Abramovitz, J., Banuri, T., Girot, T, P.O., Orlando, B., Schneider, N., Spanger-Siegfreid, E., Switzer, J. and Hammill, A. (2001). Adapting to ClimateChange: Natural Resource Management and Vulnerability Reduction.Gland, Switzerland, IUCN, World Watch Institute, IISD, StockholmEnvironment Institute, Boston.

Adger, W.N. and Kelly, P.M. (1999). Social Vulnerability to Climate Change andthe Architecture of Entitlements. Mitigation and Adaptation Strategies forGlobal Change, 4 (3-4), 253-266.

Arrow, K.J., Cline, W., Maler, K G., Munasinghe, M., Squitieri, R. and Stiglitz,J. (1996). Intertemporal Equity, Discounting and Economic Efficiency.

Bell, M., Hobbs, B., Elliott, E., Ellis, H. and Robinson, Z. (2001). An Evaluationof Multi-Criteria Methods in Integrated Assessment of Climate Policy.Journal of Multi-Criteria Decision Analysis, 10, 229-256.

Belli, P., Anderson, J.R., Barnum, H.N., Dixon, J.A. and Tan, J-P. (2001).Economic Analysis of Investment Operations: Analytical Tools andPractical Applications. World Bank, Washington, pp. 264.

Benioff, R., Guiff, S. and Lee, J. (1996). Vulnerability and Adaptation Assessments.An International Handbook, eds., Dordrecht: Kluwer Academic Publishers.

Bernthal, F., Downdeswell, E., Luo, J., Attard, D., Vellinga, P., Karimanzira,R. et al. (1990). Eds. Climate Change. The IPCC Response Strategies.WMO/UNEP, pp. 270.

Burton, I. (2000). Adaptation to Climate Change and Variability in the Contextof Sustainable Development. In: Gomez-Echeverri, L. ed. ClimateChange and Development, Yale School of Forestry and EnvironmentalStudies, pp. 153-173.

Burton, I., Kates, R.W. and White, G.F. (1993). The Environment as Hazard.New York: Guilford Press.

Fankhauser, S. (1998). The Costs of Adapting to Climate Change. GEF WorkingPaper 16.

Feenstra, J.F., Burton, I., Smith, J.B. and Tol, R.S.J. (1998). Handbook onMethods for Climate Change Impact Assessment and AdaptationStrategies. UNEP/IVM, Nairobi/Amsterdam.

GEF (2002). Operational Guidelines for Expedited Funding for the Preparationof National Adaptation Programs of Action. GEF, Washington.

Jepma, C.J., Asaduzzaman, M., Mintzer, I., Maya, R.S., Al-Moneff, M., Byrne,J., Geller, H., Hendriks, C.A., Jefferson, M., Leach, G., Qureshi, A., Sashin,W., Sedjo, R.A. and Van Der Veen, A. (1996). A generic assessment ofresponse options. In: Bruce, J.P., Lee, H. and Haites, E.F. eds., ClimateChange 1995. Economic and Social Dimensions of Climate Change,Cambridge: Cambridge University Press, pp. 225-262.

Kelly, P. and Adger, W.N. (1999). Assessing Vulnerability to Climate Changeand Facilitating Adaptation. Working Paper GEC 99-07, Centre forSocial and Economic Research on the Global Environment, Universityof East Anglia, Norwich.

Klein, R.J.T., Tol, R.S.J. (1997). Adaptation to Climate Change: Options andTechnologies. An Overview Paper. Institute for Environmental Studies,Amsterdam, pp. 33.

Klein, R.J.T., Nicholls, R.J. and Mimura, N. (1999). Coastal Adaptation toClimate Change: Can the IPCC Technical Guidelines Be Applied? In:Mitigation and Adaptation Strategies for Global Change, KluwerAcademic Publishers, 4 (3-4), 239-252.

Kuyvenhoven, A. and Mennes, L.B.M. (1985). Guidelines for Project Appraisal.Erasmus University. The Hague: Government Publishing Office.

Least Developed Countries Expert Group (2001). Guidelines for theEstablishment of National Adaptation Programmes of Actions. UNFC-CC, Bonn.

McCarthy, J.J., Canziani, O.F., Leary, N.A., Dokken, D.J. and White, K.S.(2001). Climate Change 2001: Impacts, Adaptation, and Vulnerability.Cambridge: Cambridge University Press.

Munasinghe, M. (2002). Framework for Analyzing the Nexus of SustainableDevelopment and Climate Change Using the Sustainomics Approach,Draft, Munasinghe Institute for Development (MIND).

Munasinghe, M. and Swart, R. eds. (2000). Climate Change and its Linkageswith Development, Equity, and Sustainability. Proceedings of the IPCCExpert Meeting held in Colombo, Sri Lanka. LIFE/RIVM/World Bank,

Mustafa, D. (1998). Structural Causes of Vulnerability to Flood Hazard inPakistan. Economic Geography, 74(3), 289-305.

OECD (2002). The DAC Guidelines. Integrating the Rio Conventions intoDevelopment Co-operation. OECD, Paris. pp. 104.

Orlando, B.M. and Klein, R.J.T. (2000). Taking an Ecosystem Approach toClimate Change Adaptation in Small Island States. 2nd Alliance of SmallIsland States Workshop on Climate Change Negotiations, Strategy andManagement, Samoa.

Parry, M. and Carter, T. (1998). Climate Impact and Adaptation Assessment.A Guide to the IPCC Approach, London: Earth Scan.

Ramakrishnan, P.S. (1998). Sustainable Development, Climate Change ad theTropical Rain Forest Landscape. Climatic Change, 39, 583-600.

Smit, B., Pilifosova, O., Burton, I., Challenger, B., Huq, S., Klein, R.J.T., Yohe, G.et al. (2001). Adaptation to Climate Change in the Context of SustainableDevelopment and Equity. In: McCarthy, J.J. et al. eds. Climate Change 2001:Impacts, Adaptation and Vulnerability, Cambridge: Cambridge UniversityPress, 875-912.

UNFCCC (1999). Compendium of Decision Tools to Evaluate Strategies forAdaptation to Climate Change. UNFCCC, Bonn.

Van Pelt, M.J.F. 1992. Sustainability-oriented Project Appraisal for DevelopingCountries. PhD Thesis, Rotterdam.

Wang’ati, F.J. (1996). The Impact of Climate Variation and SustainableDevelopment in the Sudano-Sahelian Region. In: Ribot, J.C., Magalhaes,A.R. and Panagides, S.S. eds. Climate Variability, Climate Change andSocial Vulnerability in the Semi-Arid Tropics, Cambridge: CambridgeUniversity Press, 71-91.

Watson, R.T., Zinyowera, M.C. and Moss, R.H. eds. (1996). Climate Change1995. Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses. Cambridge: Cambridge University Press, 879.

Willows, R.I. and Connell, R.K. eds. (2003). Climate Adaptation: Risks,Uncertainty and Decision-Making. UKCIP Technical Report, Oxford.

Winpenny, J.T. (1992). Values for the Environment. A Guide to EconomicAppraisal. London: Overseas Development Institute.


Annex A.8.1. Methods for the prioritisation and selectionof adaptation policies and measures

This annex discusses methods for selecting and prioritisingadaptation policies and measures, as well as using experts:

• Cost Benefit Analysis (CBA)• Cost Effectiveness Analysis (CEA) • Multi-criteria Analysis (MCA)• Expert judgement

Numerous textbooks exist on cost-benefit analysis, includingcost-effectiveness analysis. Sensitivity and risk analysis is ofspecial importance when dealing with the many uncertaintiesof climate change and extreme events. Reference is made to theliterature for full treatment of the methods.11

A.8.1.1. Cost-benefit analysis

CBA involves comparing costs and benefits of a measure witha view to deciding whether it is attractive to undertake anactivity (a project or a project-type adaptation measure). It isnormally applied at the country level and for estimating thecontribution of the measure to the national economy or soci-ety. However, the method can also be applied at the interna-tional or provincial level, as well as for private enterprise.Application requires making estimates of costs and benefits,involving three steps: (1) identify which costs and benefits arerelevant, (2) quantify them, and (3) give them monetary value.Although benefits are not always quantifiable and/or cannotalways be expressed in monetary values, the costing of mea-sures is possible as long as priced resources are used. The non-monetary use of scarce resources, such as utilising existingcapacity in government, should also be estimated and be takeninto account.

The method is data intensive and needs specialists. However,before a proposed measure can be entered into any plan, itsfinancial costs have to be known. Data on costs and all kinds ofparameters needed in economic analysis are normally availableat planning bureaus, ministries of planning, line ministries, etc.

Typically, professional economists perform CBA using spread-sheet software. The technical nature of the method precludesdoing more here than indicating some specific issues of impor-

tance when performing CBA. IPCC (Jepma et al., 1996) and oth-ers (e.g., Belli et al., 2001) provide guidance on this method.Techniques for introducing equity considerations into CBA areprovided in Kuyvenhoven and Mennes (1985) and Van Pelt(1992). The valuation of environmental benefits is discussed inWinpenny (1992), among others.

CBA allows analysts to optimise both the extent and the timing ofa measure. When the growth of benefits decreases with increasedintensity or extent of area of a measure, there comes a situationwhen marginal or incremental costs (MC) exceed incremental ormarginal returns (MR). At the optimum the well known conditionMC=MR applies, where the NPV (net present value) is at its max-imum.12 Postponement of a measure may result in a higher NPV.

Costing, or more generally, the valuation of costs and benefitsis an important Component of CBA, CEA and often also inMCA. It is treated in a separate section below.

A.8.1.2. Cost-effectiveness analysis

If benefits cannot be measured in a reliable manner, as is thecase often with environmental goods and services, for instance,CEA is the appropriate method. It principally involves the cost-ing of different options, which achieve the same objective, andcompares those in order to find out how a well-defined objec-tive can be reached in a least-cost way. If there are multipleobjectives, CEA can only be applied if one objective can, quan-titatively, be expressed in the other by assigning importance(weight) to the objectives to arrive at a single yardstick. This iscalled “weighted CEA”.

A.8.1.3. Multi-criteria analysis

MCA has become increasingly popular, not least in relation toenvironmental issues, including climate change (Arrow et al.,1996; Belli et al., 2001). Methods and software have proliferated.Some authors have attempted to compare different methods.13

These reviews have provided useful insights. Provisional conclu-sions on MCA are:

• Method uncertainty: Different methods produce differ-ent results and it therefore appears preferable to applyseveral MCA methods (Belli et al., abstract, p.229);

ANNEXES

11 Software programmes that may be useful for the prioritisation and selection process include the following: Multicriteria Analysis: Manual of DTLR 2001http://www.dtlr.gov.uk/about/multicriteria; HIVIEW for Windows http://www.enterprise-lsa.co.uk; DEFINITE Institute for Environmental Studies, VrijeUniversiteit, Amsterdam, The Netherlands, http://www.vu.nl/ivm@RISK http://www.palisade.com/html/risk

12 Samuel Fankhauser (1998) expresses the optimisation slightly differently. He minimises the sum of adaptation costs and residual damages. “Residual dam-ages”, when avoided by a measure, become “benefits” in CBA language. Note that, also in that reasoning, it is normally not economically efficient to avoidall (residual) damages.

13 Belli et al., (2001) distinguish three groups of MCDM methods: weighting methods, deterministic ranking methods and uncertainty ranking methods.


• Ease of manipulation coupled with subjectivity andlack of transparency has contributed to lack of confi-dence in MCA methods. Some recommend simplermethods, preferably without use of computer software;

• MCA is very useful for structuring problems and deci-sions, not necessarily for solving problems (holisticassessments are preferred for ultimate decisions).

The ingredients of MCA are objectives, alternative measures/interventions, criteria (or attributes), scores that measure or valuethe performance of an option against the criteria, and weights(applied to criteria). Defining objectives and formulating differ-ent options is not different from CBA or CEA. The differencelies in the selection of criteria and their weights. As indicatedabove, these are judgmental elements. For the determination ofweights, procedures exist that more or less guarantee that the setof weights is consistent. For example, the computer MCA model“DEFINITE” contains a separate routine (pair-wise comparison)to arrive at a consistent set of criteria. That model further allowsusing different ways of MCA, from simple to quite sophisticat-ed, and includes a routine for CBA (Annex A.8.3).

A major task is determining the scores (or effects), i.e., assess-ing the impact of alternative measures on the different criteria.Assessing causal relationships between measures and effects isa matter of research. This may in practice be the most impor-tant task, since there is no method that can make up for unreli-ability of input data.

The selection of a set of criteria is subject to a number of pitfalls.Probably the most serious danger is overlap (double counting) orinterdependency. Another danger is that only those criteria areselected to which effects can easily be attributed. Health and bio-diversity are criteria that may fall victim to the difficulties of esti-mating and attributing effects. Too many criteria may be takeninto account, leading to a “splitting bias”. According to Van Pelt(1992), MCA is most reliable if the number of alternative optionslies between three and eight, the number of criteria does notexceed seven, the impact can be quantified, and if different MCAtechniques give comparable outcomes.

On the positive side are:

• Apart from forcing the user to frame the problem (seeabove), MCA provides a checklist of data requiredand the sensitivity of inputs and the result can easilybe analysed. In a way, MCA guides the data collectionprocess;

• MCA is particularly suitable for use in a “participativesetting” (especially in determining the relevance andthe weights of criteria) and so allows stakeholder par-ticipation in a systematic way through the variousstages of the APF, such as scoping of adaptations,problem definition, determining relevance of inputdata and feeding back results to stakeholders.

In Annex A.8.3, a hypothetical example is given to furtherexplain the selection/prioritisation procedure using MCA.

A.8.1.4. Costing

Costing is of importance for all methods, and for each, thesame principles apply. Basically, three steps are involved:

• Identification: determining which costs and benefitsare relevant.

• Quantification: measuring inputs and effects in terms of,say, labour days, tons of produce, number of casualties.

• Valuation: pricing the in- and outputs.

There is, in this respect, no difference between costs and bene-fits, defined respectively as the decrease and increase of scarceresources. Often benefits are a decrease of costs that would beincurred in the absence of the project.

Financial costs

An estimate of financial costs is the starting point for costingeconomic or social costs. Financial costs are the outlays forthe project to be made by the agency implementing the pro-ject. Economic or social costs are losses of scarce resourcesfrom the point of view of the whole society. The two notionsrarely coincide and corrections on financial cost will have tobe made.

Prices of production factors (labour, capital, expertise) and ofgoods/services are often distorted. Main sources of distortionare indirect taxes/subsidies and other deliberate governmentpolicies and mal-functioning of markets. Corrections are usuallynecessary for:

• taxation/subsidisation;• wages;• discount rate (interest, the price of capital);• foreign currency (exchange rate).

A cost-benefit exercise presumes that, with regard to benefits,the changes in vulnerability can be measured. Issues of estima-tion and uncertainty make this difficult. On both fronts, how-ever, progress is made (references, especially Belli et al., andWinpenny). Here the focus is on costing as this is a complexexercise in itself.

Social costs

While financial costs are fairly straightforward, social costs, asmentioned earlier, need to account for market distortions, trans-fer payments, and external effects.


Market distortions: A resource or activity may be zero-pricedin its present use. For instance, land leased by the governmentfor agriculture is now being used by a project to plant forestsin watersheds to prevent soil erosion, and mangroves to act asstorm breakers. Or, the land may be priced at the lease value.That value, however, may be too low relative to the value ofthe crops produced. Revaluation is often needed to reflect thereal value of the resource or related activity to the communi-ty. A guiding concept here is that of “opportunity costs”.When applied to labour in the project, the reasoning is that thetrue cost of labour is the added value forgone (e.g., in termsof rice produced). Applying the opportunity costs reasoningto material inputs leads to the use of “border prices” or long-run world market prices. The “shadow” exchange rate is therate that would prevail in the absence of undue protection(more than of trading partners) and when the rate is left tofluctuate freely. The estimation is specialised work. There isa debate on the discount rate to use for investments with avery long time horizon and/or dealing with irreversibleeffects. Some argue for a lower rate in those cases in order tofoster acceptance of such projects, often environmental pro-jects (refer also to Adger and Kelly, 1999).

Transfer payments: Taxes and subsidies are costs/income forthe project, but for the community as a whole these are meremoney transfers and should therefore be eliminated in eco-nomic costing. Note, however, that a user charge (such as a levyfor irrigation water or a road user charge) represent a use of aresource (preferably, at least from the point of economic effi-ciency) equal to the actual cost made for the provision of thewater or the road system.

External effects: The use of the resource may entail additionalcosts or benefits outside the measure or project being consid-ered. These may or may not be quantifiable. For instance,afforestation may generate additional recreational or biodiver-sity benefits. Similarly, mangrove plantation may improvespawning conditions for aquatic species, improve biodiversityand provide timber, fuelwood and fodder. These are externalbenefits. CO2 produced during construction of (protective)infrastructure would be an external cost14. Also additionalmonetary and non-monetary benefits may be associated withan adaptation project/activity. For instance, in coastal zones,water conservation measures could increase soil productivityby raising the water table and reducing salinity. Negative costs(external benefits) occur when such secondary benefits/co-ben-efits/joint benefits more than offset the additional investment inadaptation. Win-win or no-regrets is another term applied tosuch projects/activities. When faced with competing prioritiesand scarce financial resources, identifying such projectsbecomes imperative. Conserving natural resources is a classiccase, offering a range of social and environmental co-benefits,such as biodiversity conservation, enhanced sink capacity,poverty alleviation and reduced demand for international assis-tance (Abramovitz et al., 2001).

Incremental cost

Incremental costs refer to marginal costs increasing in concretesteps. In the climate change community, the term incremental costas used, inter alia, by the Global Environment Facility (GEF) isdefined as the additional cost a country incurs when undertakinga climate mitigation project, compared with the social cost of theactivity the project substitutes and that has no provisions for mit-igation of green-house gases.

While the IPCC TAR applies the incremental cost criteria acrossmitigation and adaptation projects, its relevance for adaptationprojects is less obvious. Such projects have no global rationale,except when they intersect with mitigation projects. On the otherhand, a case for applying such criteria to funding adaptation pro-jects can be made. For instance, a dam wall is raised to addressthe threat of increased flooding but also stores more water forlater irrigation releases. Applying these criteria means the latterbenefit would need to be netted out of the project cost. This is animportant consideration, especially in view of the limited avail-able funding for adaptation.

Time discounting

The present value of a future cost stream is defined as:

Present cost = (future costs)/(1+discount rate)t

Where the exponent “t” refers to the time stream of costs. Thediscount rate in the economic analysis can range from a low“ethical” rate, based on social considerations, to a rate whichreflects the opportunity cost of capital. Discount rates varybetween developed and developing countries and, basically, thescarcity of capital should determine the choice of the rateapplied.

As the discount rate is very important for the outcome and oftendifficult to estimate, it is often made subject to sensitivity analy-ses to determine how sensitive the results are to the choice of thediscount rate. Also, when estimating the time stream of cost it isnecessary to spell out clearly the assumptions underlying theforecasts and how these assumptions are used to generate theforecasts, including linkages and feedback effects.

Implementation costs

Estimating direct project costs is not sufficient. It is also importantto assess the institutional, economic and technical barriers toimplementing the project because additional costs (financial or inkind) are involved in removing them. The required changes maybe institutional (e.g., improving adaptive research capacity) or eco-nomic (e.g., establishing markets and incentives for new products).The cost of these changes ought to be added to the project.

14 CO2 emission is an external cost at global level (leads to climate change), and may – in a national-economic analysis – be negligible, especially for smalldeveloping countries.


Arguably, barriers to implementation may be less of a concern inthe case of adaptation projects, as such projects tend to be main-streamed into the policy system – for instance, early warning andmitigation of damages in relation to floods/droughts/cyclones,afforestation, water conservation and health interventions.Admittedly, there are different degrees of mainstreaming.

Combining methods

As discussed above, a single method will not normally suffice.Often a combination of methods may be called for and alsoresults of different methods may be checked against each other.The MCA model DEFINITE provides four MCA methods andCBA/CEA.

A.8.1.5. Using an expert panel

It may also be that data unavailability or the complexity of theproblem suggest the use of expert judgment. Employing a panelof experts may aim at taking a decision or at producing suitableinformation for decision making. A structured way to engageexperts is the DELPHI method. It involves sending question-naires to experts (rather than getting them in one meeting), col-lating the answers, and feeding those back to the experts and/orsend a second questionnaire. An important ingredient of DEL-PHI is that experts give their opinions independently and anony-mously. The results are given in the form of a statistical analy-sis of answers. DELPHI is mostly applied to forecasting (e.g., oftechnological development), but may be of some use here, forinstance for formulating adaptation options.

A.8.1.6. Handling uncertainty and risk

Climate change is a process that is characterised by a numberof uncertainties relative in particular to the magnitude, timingand nature of the changes. Decision makers are more familiarwith processes/problems that are not subject to this degree ofuncertainty. To take into account this situation, various methodscan be used.

A common method in project appraisal is sensitivity analysis.Main inputs in the analysis (such as certain cost and/or bene-fits, the discount rate, etc.) are varied to see how sensitive theoutcome is to these changes. A practical approach is to deter-mine “switching values”, i.e., those values of major inputs,either alone or in combination with others, that render an activ-ity uneconomic. A similar procedure can be used in (comput-erised) MCA analysis.

Risk analysis uses Monte Carlo simulation on key inputs in theanalysis. The analyst has to determine the probability distribu-tion (normal, skewed, etc.) of the occurrence (say of an increasein cost, or number or intensity of extreme events) and the possi-

ble co-variance between these inputs.15 The computer model,using a random number generator, makes a large number of runsto determine the (average) outcome. If a probability (or combi-nations) is an input into the analysis, the output, naturally, is alsoa probability distribution (e.g., of the NPV or the rate of return).There are commercial computer packages that perform the work,(Burton, 2000). Also in MCA models, routine risk analysis canbe built in (Burton et al., 1993).

Another possibility to deal with uncertainty is scenario develop-ment. This method corresponds to the analysis proposed in TP6and to climate change scenarios developed by IPCC. For a sys-tematic analysis the likelihood of a situation occurring and itsprobability distribution could be used as input in risk analysis.

Annex A.8.2. Linking climate change and sustainabledevelopment policies

Munasinghe (2002) developed a method to link climate changeand sustainable development.

Integrated assessments – The Action Impact Matrix (AIM)

There is a two-way linkage between climate change and sus-tainable development. Future development paths (hence theneed for charting socio-economic scenarios) will determinenot only projected greenhouse gas (GHG) emissions and theseverity of climate change, but also the adaptive and mitigativecapacity available to mount an effective response strategy.Conversely, climate change will have significant impacts onthe three main elements of sustainable development (econom-ic, social and environmental).

This dynamic interaction should become a consideration indevelopment cooperation. Integrated sustainable developmentand climate change policies should account for the powerfuleconomy-wide reforms in common use – including both sec-toral and macroeconomic adjustment policies, which havewidespread effects throughout the economy.

The highest priority needs to be given to policies that promoteall three elements of sustainable development (economic,social, environmental). This is especially in recognition of thefact that there are sustainable development issues, which affecthuman welfare more immediately – such as hunger, malnutri-tion, poverty, health, and pressing local environmental issues.With other policies, trade-offs among different objectives needto be analysed. Economy-wide policies that successfullyinduce growth, could also lead to environmental and socialharm, unless the macro-reforms are complemented by addi-tional environmental and social measures.

The AIM provides a way to link integrated sustainable devel-opment and climate change policies explicitly. It can help find“win-win” policies, which not only achieve conventional

15 See Belli et al. for a concise treatment of risk analysis.


Activity/Policy(PRSPs,NSSDs)

Economicobjectives

Impacts on key sustainable development issues

Land degradation.Biodiversityloss

Water scarcityand pollution.Adverse healthimpacts

Air emissions.Adverse healthimpacts

Other socialeffects

Institutionalimpacts

Vulnerability(socio-economic andbiophysical)

Macroeconomicand sectorpolicies

Macroeconomicand sectorimprovements

Positive impacts Negative impacts Indeterminate impacts (I)

Exchange rate Improve tradebalance andeconomicgrowth

Deforest openaccess areas

Water pollution (I)Adverse healthimpacts in lowincome areas

Air pollution (I)Adverse healthimpacts in lowincome areas

Forced migration toother areas

Increase vulnerability

Water pricing andmanagement

More efficientwater use andeconomic efficiency

Reducedwaterlogging/salinity

Water use efficiency

Improvedaccess to water for poor farmers

Integratedwater anddrainage management

Reduce vulnerability

Energy pricing andmanagement

Increase energy use efficiency

Reduced biomass use

Reduced airpollution.Lower health risks


Comple-mentarymeasures

Socio-econom-ic and marketgains

Enhance positive impactsMitigate negative impacts

Market based Pollutioncharge

Emissioncharge


Non-marketbased

Institute property rights

Voluntary compliance

Voluntary compliance

Amendingenvironmentallaws/regulations


Adaptationprojects

Reducevulnerability

Investment decisions guided by broader policy and institutional frameworkPositive impactsNegative impactsIndeterminate impacts (I)

Re-forestation/aforestation

Increase sinkscapacity.Reduce soilerosion anddownstreamsedimentation

Reduce flooding

Provide fuel,timber and fodder to poorcommunities.Reduce damagecaused byflooding


Raising damwalls

Seepage,inundateforests

Floodingeffects (I)

Potentialhydropower.Reduced airemissions

Socio-econom-ic effects offlooding (eco-nomic losses,displacement,mortality) (I).

Vulnerabilityeffects (I)

Drought miti-gation (relief,early warning,infrastructureand services)

Reduce pressure onland (crop and grazing)

Improvedsocio-econom-ic conditions


Note: PRSP = Poverty reduction strategy paperNSSD = National strategy for sustainable development

Source: Munasinghe and Swart (2000)

Table A-8-2-1: A simple version of the Action Impact Matrix


macroeconomic objectives (like growth), but also make localand national development efforts more sustainable, and addressclimate change issues. AIM demonstrates in practical and qual-itative terms that economic growth, social justice and environ-mental sustainability can co-exist.

The rows list the main development interventions (both poli-cies and projects), while the columns indicate key sustainabledevelopment issues and impacts (including climate change vul-nerability). Thus the elements or cells in the matrix help to:

• Identify explicitly the key linkages.• Focus attention on methods of analysing the most

important impacts.• Suggest action priorities and remedies.

At the same time, the organisation of the overall matrix facili-tates the tracing of impacts, as well as the coherent articulationof the links among a range of development actions – both poli-cies and projects.

A simple version of the AIM is presented in Table A-8-2-1.

Annex A.8.3. A hypothetical example of the use of multi-criteria analysis

This illustration of MCA uses a hypothetical example a situa-tion, which may typically exist in the wet season in Bangladesh.The steps normally taken in a MCA analysis are as follows:

1) Problem definition: Because of rising sea level, higherintensity of precipitation and increased run-off inupstream areas, rain and smelt water reaches Bangladeshin a shorter period than before and also drains less easi-ly. Floods, as Bangladesh has been experiencing for along time, are thus getting worse.

2) The objective of the intervention is to get rid of super-fluous water in order to safeguard agricultural produc-tion, to avoid the spread of waterborne diseases, and toavoid damages to buildings, nature, infrastructure, etc.(called “environment” in the example).

3) The criteria used to measure effects are: (a) agricultur-al production, (b) health, (c) expected damage of theenvironment, and (d) the cost of the intervention.

4) The following are considered as alternative inter-ventions: (a) installing pumps at strategic sites, (b)improving the existing drainage infrastructure, (c)organising manual labour at a big scale (not unusu-al in Bangladesh). An alternative option is always todo nothing (bear the losses).

5) Estimating effects in a reliable manner naturally is ofparamount importance. This is the area where riskanalysis is especially valuable. It is here assumed thatthere is insufficient data at this stage to perform eitherCBA or CEA. For the study area, however, there arerough estimates of the extent and duration of floods

Table A-8-3-2: Scores standardised (0-1 scale), weighted summation and ranking

Cost Effect Health Environment Weightedsummation

Ranking

Pump 0.22 1.00 1.00 0.00 0.56 2

Infrastructure 0.11 0.80 0.80 1.00 0.68 1

Labour 0.00 0.30 0.30 1.00 0.40 3

Bear losses 1.00 0.00 0.00 0.33 0.33 4

Weight 0.25 0.25 0.25 0.25 1.00

Table A-8-3-1: Scores on Criteria

Cost (million $) Effect(million HA days)

Health(million DALYs)

Environment cost(million $)

Pump -700 1000 10 -70

Infrastructure -800 800 8 -10

Labour -900 300 3 -10

Bear losses 0 0 0 -50


that could be avoided, of the daily adjusted life years(DALYs) (Belli et al., 2001) that could so be gained, ofthe damage done to the environment (in money terms)and of the costs of the different interventions (also inmoney terms).

6) The last step is to give weights to the different criteria.

All steps lend themselves to stakeholder participation, espe-cially Steps 3, 4 and 6. Under Step 4, traditional coping mech-anisms would be brought in, and under Step 6, the preferencesof people affected by the floods and the measures to avoidthem. Table A-8-3-1 gives the basic data and Table A-8-3-2 theresults in terms of ranking after the effects (expressed in vari-ous units) have been standardised by scaling them (0 – 100 or0-1 scale) and weights have been assigned to the criteria.

The data was input in a spreadsheet (Table A-8-3-1) and thecalculations (standardisation and summation of contribution)are made in Table A-8-3-2. When the calculations are formu-lated in the spreadsheet, performing sensitivity is easy.16 Itwould be logical to investigate the sensitivity of the effects(scores in Table A-8-3-1) and of the weighting (Table A-8-3-2)on the outcome (ranking). If pumping would be less expensive(for instance, only 400) and infrastructure more expensive(950) the two alternatives would get the same ranking. If costwould be given a weight of 0.45 and environment of 0.05, thenpumping becomes the better alternative. Also risk analysiscould be done on the scores (Burton, 2000), but first every

effort should be done to improve the estimates of effects. MCAcan be done also using computerised models. Both HIVIEWand DEFINITE (Annex A.8.4) support weighted summation, asdone above. Doing sensitivity using a computerised MCAmodel is relatively easy.

Both models are easy to apply. HIVIEW does sensitivity,allows relative and absolute scaling, and accepts inputs ofscores in various forms (numbers, but also “yes” and “no”).HIVIEW can also present the structure of weighting graphical-ly. DEFINITE is a full-fledged decision support programme. Itincludes four different MCA methods, CBA and graphical eval-uation methods. It allows all formats for inputs, including +,++, -, —. DEFINITE leads the analyst through rounds of inter-active assessments of options, weights, scores, etc.; there is aroutine to check internal consistency of the weight set usingpair-wise comparison. Reporting is in text and numbers, butgraphs can also be produced, as shown here in Figure A-8-3-1.Some training in the models is highly recommended beforeusing them to propose important decisions.

The danger of the programme (as with all MCA) is that theemphasis comes on the method, rather than on the (hard) workto develop estimates of costs and benefits of options.

Annex A.8.4. Useful internet addresses

UNFCCC Secretariat: www.unfccc.intUNITAR: www.geic.or.jp/cctrainIPCC: www.ipcc.ch/pub/tar/wg2/069.htmWorld Bank: www.worldbank.orgStratus Consulting: www.stratusconsulting.comStockholm Environment Institute (SEI): www.tellus.comAssessments of Impacts and Adaptations to Climate Change(AIACC): www.start.org/Projects/AIACC_Project/aiacc.htmlUnited Kingdom Climate Impacts Programme (UKCIP):www.ukcip.org.uk

Models

HIVIEW for Windows http://www.enterprise-lsa.co.uk; DEFI-NITE Institute for Environmental Studies, Vrije Universiteit,Amsterdam, The Netherlands, http://www.vu.nl/ivm@RISKhttp://www.palisade.com/html/risk

16 Note that the scaling of scores is always done on the highest/lowest score in a column and that this order can change in the sensitivity analysis.

infr

astr

uctu

re

pum

p

labo

ur

bear

loss

es

Res

ult

1

0

0.68

0.56

0.400.33

Figure A-8-3-1: MCA 1: Weighted summation {interval;direct (cost: 0.5)}